OpenAI CEO Sam Altman.
Image: Mike Coppola / Getty Images North America / Getty Images via AFP
In 2024, OpenAI took steps to remove the oversight that its non-profit board has on the organisation’s profit-seeking subsidiary. The company had seen the sacking of co-founder and CEO Sam Altman in November 2023 and then his equally dramatic return in March this year. Altman is widely seen as the prime driver of the for-profit shift.
Microsoft, with $13 billion invested in OpenAI, is its biggest backer. And Altman is moving to remove a rule governing OpenAI that blocks the big tech backer from accessing its advanced AI models if it achieves artificial general intelligence.
Financial Times noted that this would allow OpenAI continued access and investments from Microsoft, as it seeks to become a regular for-profit tech company. Here are five takeaways looking back.
OpenAI is restructuring itself to become a for-profit company and co-founder and CEO Sam Altman is to get a 7 percent equity stake.
OpenAI was founded in 2015 as a non-profit research organisation to build artificial intelligence (AI) for the benefit of humanity. The company created a for-profit subsidiary in 2019, faced with the high costs of developing its large language models. Altman had thus far declined to take a stake in the company.
Meanwhile, several top leaders at the company have left, including co-founder John Schulman, CTO Mira Murati and Miles Brundage, a senior advisor on artificial general intelligence. Murati is said to be starting an AI venture of her own.
In talks to remove clause blocking Microsoft’s access to AGI
OpenAI is also in talks to remove a provision that would cut Microsoft off from its advanced models once it achieves ‘artificial general intelligence’ (AGI), Financial Times reports.
Currently, if AGI is achieved, Microsoft’s access to OpenAI’s technology would be void, as the startup’s non-profit board would own the technology.
However, OpenAI is exploring removing this stipulation to maintain Microsoft’s continued investment and access, which could help secure billions for future AI development. The clause was initially meant to prevent AGI from being misused commercially.
And as OpenAI becomes a public benefit corporation, discussions are ongoing about the future role of Microsoft and other investors, according to FT.
Raises $6.6 billion at a valuation of $157 billion
OpenAI announced $6.6 billion in new funding in October, its biggest since Microsoft’s $10 billion investment in January 2023. This latest round raised its valuation to $157 billion, the company said in a statement.
The funding round was led by Thrive Capital, which had committed to investing $1.2 billion, with the option of investing another $1 billion next year, according to Reuters.
Other big contributions came from Microsoft, which is said to have invested a little less than $1 billion, according to reports citing the Wall Street Journal. SoftBank invested $500 million.
Additional notable investors included Khosla Ventures, Fidelity Management, Tiger Global Management, Ark Investment Management and Nvidia. Reuters reported that the company was on track to generate $3.6 billion in revenue this year while losses rise to over $5 billion. OpenAI is projecting revenue of $11.6 billion next year.
Rival xAI, started by Elon Musk, has raised around $6 billion in equity financing. The funding comes at a time xAI is looking to increase its AI footprint by expanding its Memphis, Tennessee, supercomputer to house at least one million graphics processing units.
Also read: Five takeaways from Nvidia’s rise and rise in 2024
Hits 300 million weekly users and counting
OpenAI has reached 300 million weekly active users. CEO Sam Altman shared this milestone at The New York Times’ DealBook Summit earlier this month, with plans to target 1 billion users within the next year, CNBC reported. Only in August that number was 200 million, which itself was a 2X growth from November 2023, according to reports from Axios and The Verge.
The company faces strong competition from rivals such as Anthropic, xAI, Perplexity and search giant Google and other big tech companies including Meta, which offers its own large language models. Big tech capital expenditure on generative AI will hit $1 trillion in the next few years, Goldman Sachs estimated in August.
OpenAI has also hired its first CMO, Kate Rouch from Coinbase, signalling a stronger focus on marketing, CNBC points out. The company’s ChatGPT is becoming ubiquitous, with its basic features available for free to anyone with a net connection. OpenAI’s deal this year to bring its tech to Apple’s devices will only accelerate its adoption through next year.
Earlier this year, the company also recruited its first employee in India, Pragya Misra, to fill a public policy and partnerships role.
Is Orion the next big model, and meanwhile, a $200 subscription
OpenAI has denied reports suggesting it will release a new AI model, code-named Orion, by December, according to a TechCrunch report. The company clarified there were no plans to release Orion this year, though other technologies will be launched, according to the report.
Some reports, including one from The Verge, had earlier suggested that Orion was expected to be OpenAI’s next major model, and that it would debut by December, with trusted partners gaining early access.
Orion is said to be an advanced version of GPT-4, trained partly using synthetic data from OpenAI’s reasoning model, o1. Meanwhile, OpenAI has launched a version of ChatGPT at $200 per month, which can be used in engineering fields and for research, according to a Reuters report.
The new tier, called ChatGPT Pro, will be in addition to OpenAI’s existing subscriptions of ChatGPT Plus, Team and Enterprise. On December 9, the company also announced that Sora, its video generating AI model was moving out of research preview into a standalone product for Plus and Pro subscribers.